Thrombocytopenia (platelet count <150,000/mcL) is a common problem among sick newborn infants, affecting 20-35% of all patients admitted to Neonatal Intensive Care Units (NICUs). In approximately 25% of these cases (20,000 to 35,000 neonates/ year in the USA), the platelet count reaches levels that prompt the administration of platelet transfusions. However, a direct correlation between degree of thrombocytopenia and bleeding risk has not been demonstrated, and the only platelet transfusion trigger study conducted in neonates was limited to very-low-birth weight (VLBW; <1500 g) infants in the first week of life, and excluded those with platelet counts <50,000/mcl. As a consequence of this lack of scientific evidence, there is great variability in transfusion practices among NICUs and among individual neonatologists. In addition, preliminary data suggest that 30-60% of platelet transfusions in the NICU are administered to non-bleeding neonates with platelet counts between 50,000 and 150,000/mcl, despite the lack of a documented favorable risk/benefit ratio at those platelet counts. It is therefore evident that a randomized controlled trial in neonates is needed to help neonatologists make safe platelet transfusion decisions, while preserving a valuable blood bank resource that is in limited supply. As a first step toward conducting such a study, we established a network of investigators and NICUs with a strong focus on neonatal transfusion medicine. Next, we recognized the need to obtain critical preliminary data to help us design a safe and scientifically sound neonatal platelet transfusion trial. The present proposal was designed to fulfill this requirement using existing data from a large historic cohort. We hypothesized that: 1) platelet counts <30,000/mcL will be associated with an increased risk of bleeding in near-term or term neonates, as well as in VLBW infants >7 days old ("lowrisk group"); and 2) Platelet counts <50,000/mcL will be associated with an increased risk of intraventricular hemorrhage (IVH) in VLBW infants <7 days old ("high risk group"). To test these hypotheses, we formulated the following two Specific Aims: 1) To establish the correlation between different degrees of thrombocytopenia and clinically significant bleeding in near-term and term infants, as well as in VLBW neonates >7 days old;and 2) To establish the correlation between different degrees of thrombocytopenia and incidence of IVH in VLBW infants <7 days old.